Chromatin is a dynamic protein-DNA complex which is modulated by post-translational modifications. These modifications, in turn, regulate cellular processes such as gene transcription and replication. Key chromatin modifications include the acetylation and deacetylation of nucelosomal histone proteins. Acetylation is catalyzed by histone acetylases (HATs), whereas deacetylation is catalyzed by deacetylases (HDACs or HDAs). HDACs catalyze the removal of acetyl groups from the N-termini of histone core proteins to produce more negatively charged chromatin. This results in chromatin compaction, which shuts down gene transcription. In addition, inhibition of HDACs results in the accumulation of hyperacetylated histones. This, in turn, is implicated in a variety of cellular responses, including altered gene expression, cell differentiation, and cell-cycle arrest (see, generally, S. G. Gray et al., 2001, Exp. Cell Res. 262(2):75-83, and U.S. Pat. Nos. 6,110,697 and 6,068,987 to Dulski et al.).
The HDAC gene family is composed of two distinct classes. Class I HDACs are related to the yeast transcriptional regulator, RPD3. Class II HDACs include a subgroup of proteins containing a C-terminal catalytic domain as well as a separate N-terminal domain with transcriptional repression activity. Class III HDAC proteins are related to the yeast sir2 protein and require NAD for activity. Class I HDACs are predominantly nuclear, whereas class II HDACs are transported between the cytoplasm and nucleus as part of the regulation of cellular proliferation and/or differentiation (reviewed in S. Khochbin et al., 2001, Curr. Opin. Genet Dev. 11(2):162-6).
The best characterized substrates for HDACs include histone or histone-like peptide sequences containing N-terminal lysines. However, non-histone HDAC substrates have also been identified, including several transcription factors. Non-histone substrates for HDACs include p53, androgen receptor, LEF1/TCF4 (B. R. Henderson et al., 2002, J. Biol. Chem., published online on May 1, 2002 as Manuscript M110602200), GATA-1, and estrogen receptor-alpha (reviewed in D. M. Vigushin et al., 2002, Anticancer Drugs 13(1):1-13). For these substrates, deacetylation has been shown to regulate DNA/protein interactions or protein stability. Such molecules may therefore represent therapeutic targets of HDACs. Importantly, the histone deacetylase function of HDACs represses transcription by removing the acetyl moieties from amino terminal lysines on histones, thereby resulting in a compact chromatin structure. In contrast, the non-histone deacetylase function of HDACs can either repress or activate transcription.
There has been considerable interest in modulating the activity of HDACs for the treatment of a variety of diseases, particularly cancer. Several small molecule inhibitors of HDAC have shown anti-proliferative activities on a number of tumor cell lines and potent anti-tumor activity in pre-clinical tumor xenograft models, most recently, CBHA (D. C. Coffey et al., 2001, Cancer Res. 61(9):3591-4), pyroxamide, (L. M. Butler et al, 2001, Clin. Cancer Res. 7(4):962-70), and CHAP31 (Y. Komatsu et al., 2001, Cancer Res. 61(11):4459-66). Several inhibitors are presently being evaluated as single agents and in combination regimens with cytotoxic agents for the treatment of advanced malignancies (reviewed in P. A. Marks et al., Curr. Opin. Oncol. 2001 November; 13(6):477-83). Thus, HDAC inhibitors are being developed as anti-tumor agents, as well as agents useful for gene therapy (McInerney et al., 2000, Gene Ther. 7(8):653-663).
Small molecule inhibitors of HDAC activity that have undergone extensive analysis include trichostatin A (TSA), trapoxin, SAHA (V. M. Richon et al., 2001, Blood Cells Mol. Dis. 27(1):260-4), CHAPs (Y. Komatsu et al., 2001, Cancer Res. 61(11):4459-66), MS-27-275 (reviewed in M. Yoshida et al., 2001, Cancer Chemother. Pharmacol. 48 Suppl. 1:S20-6), depsipeptide (FR901228; FK228; see, e.g., V. Sandor et al., 2002, Clin. Cancer Res. 8(3):718-28), and CI-994 (see, e.g., P. M. LoRusso et al., 1996, New Drugs 14(4):349-56; S. Prakash et al., 2001, Invest. New Drugs 19(1):1-11). Trichostatin A and trapoxin have been reported to be reversible and irreversible inhibitors, respectively, of mammalian histone deacetylase (Yoshida et al, 1995, Bioassays, 17(5):423-430). Trichostatin A has also been reported to inhibit partially purified yeast histone deacetylase (Sanchez del Pino et al., 1994, Biochem. J., 303:723-729). Moreover, trichostatin A is an antifungal antibiotic and has been shown to have anti-trichomonal activity and cell differentiating activity in murine erythroleukemia cells, as well as the ability to induce phenotypic reversion in ras-transformed fibroblast cells (see e.g. U.S. Pat. No. 4,218,478; and Yoshida et al., 1995, Bioassays, 17(5):423-430, and references cited therein). Trapoxin A, a cyclic tetrapeptide, induces morphological reversion of v-sis-transformed NIH/3T3 cells (Yoshida and Sugita, 1992, Jap. J. Cancer Res., 83(4):324-328).
The therapeutic effects of HDAC inhibition are believed to occur through the induction of differentiation and/or apoptosis through the up-regulation of genes such as the cyclin dependent kinase inhibitors, p21 and p27 (see, e.g., W. Wharton et al., 2000, J. Biol. Chem. 275(43):33981-7; L. Huang et al., 2000, Mol. Med. 6(10):849-66). Although known HDAC inhibitors are efficacious as anti-tumor agents, they are also associated with toxicity (see, e.g., V. Sandor et al., 2002, Clin. Cancer Res. 8(3):718-28). Such toxicity is believed to be caused by a non-selective mechanism of targeting multiple HDACs. Despite the potent anti-tumor activity of HDAC inhibitors, it is still unclear which HDACs are necessary to produce an anti-proliferative response. Furthermore, little progress has been made in comparing the HDAC gene expression profiles in tumor versus normal cells. Differential HDAC expression may underlie the tumor-selective responses of HDAC inhibition. In addition, a cellular growth advantage may be conferred by the expression of particular HDACs. Therefore, there is a need for further insight into the consequences of selective HDAC inhibition, or activation.